Predicting turnout on Tuesday

I wrote a post before the primary election on predicting turnout. That was quite a bit more involved since there was the issue of the new summer date to consider. Predicting turnout this Tuesday should be a little more straight forward since there are not really any unknown variables to consider, unless you think there will be a GOP surge.

I'm not discounting that possibility, I just don't think that it will have nearly the same effect in Minnesota where we already turnout at a high rate.

Here is a graph of turnout % in Minnesota mid-term elections since 1950.



As you can see there was a period between 1970 and 1994 in which there was tremendous volatility in turnout, dropping below 50% a few times and hitting over 60% only once. Over the last three mid-term's however, the trend has been a little more stable.

Instead of using the median, I'm going to use the average of the last three years for the turnout prediction. I'm doing this because this race would seem to roughly fit into the same paradigm of races that we've had in the last three elections.

This is a completely subjective decision and one I'm making more on the basis of intuition than anything else. Using the median turnout of 60% might be the correct thing to do, but I suspect that will be an underestimation.

So now that I have a turnout estimate, I just need to get an eligible voter number. In Minnesota, unlike most other states, you can't just go with the number of registered voters because we have same day registration, part of the reason we have such high turnout. So instead we have to figure out an eligible voter number.



As you can see the increase in the number of eligible voters from mid-term to mid-term has been rather stable over the last eight or so cycles, so we can be fairly confident that making an extrapolation from the number of eligible voters in 2006 will get us close.

The median increase in the number of eligible voters over that time was 3.8%. The number of eligible voters in 2006 was 3,667,707. So by doing some multiplication we arrive at an estimate of 3,807,080 eligible voters in 2010. Here than is an estimated turnout based on a couple different turnout scenarios, my prediction is in bold.